Global Position Aware Group Choreography using Large Language Model

Dance serves as a profound and universal expression of human culture, conveying emotions and stories through movements synchronized with music. Although some current works have achieved satisfactory results in the task of single-person dance generation, the field of multi-person dance generation remains relatively novel. In this work, we present a group choreography framework that leverages recent advancements in Large Language Models (LLM) by modeling the group dance generation problem as a sequence-to-sequence translation task. Our framework consists of a tokenizer that transforms continuous features into discrete tokens, and an LLM that is fine-tuned to predict motion tokens given the audio tokens. We show that by proper tokenization of input modalities and careful design of the LLM training strategies, our framework can generate realistic and diverse group dances while maintaining strong music correlation and dancer-wise consistency. Extensive experiments and evaluations demonstrate that our framework achieves state-of-the-art performance.

JamendoMaxCaps: A Large Scale Music-caption Dataset with Imputed Metadata

We introduce JamendoMaxCaps, a large-scale music-caption dataset featuring over 200,000 freely licensed instrumental tracks from the renowned Jamendo platform. The dataset includes captions generated by a state-of-the-art captioning model, enhanced with imputed metadata. We also introduce a retrieval system that leverages both musical features and metadata to identify similar songs, which are then used to fill in missing metadata using a local large language model (LLLM). This approach allows us to provide a more comprehensive and informative dataset for researchers working on music-language understanding tasks. We validate this approach quantitatively with five different measurements. By making the JamendoMaxCaps dataset publicly available, we provide a high-quality resource to advance research in music-language understanding tasks such as music retrieval, multimodal representation learning, and generative music models.

Designing Neural Synthesizers for Low Latency Interaction

Neural Audio Synthesis (NAS) models offer interactive musical control over high-quality, expressive audio generators. While these models can operate in real-time, they often suffer from high latency, making them unsuitable for intimate musical interaction. The impact of architectural choices in deep learning models on audio latency remains largely unexplored in the NAS literature. In this work, we investigate the sources of latency and jitter typically found in interactive NAS models. We then apply this analysis to the task of timbre transfer using RAVE, a convolutional variational autoencoder for audio waveforms introduced by Caillon et al. in 2021. Finally, we present an iterative design approach for optimizing latency. This culminates with a model we call BRAVE (Bravely Realtime Audio Variational autoEncoder), which is low-latency and exhibits better pitch and loudness replication while showing timbre modification capabilities similar to RAVE. We implement it in a specialized inference framework for low-latency, real-time inference and present a proof-of-concept audio plugin compatible with audio signals from musical instruments. We expect the challenges and guidelines described in this document to support NAS researchers in designing models for low-latency inference from the ground up, enriching the landscape of possibilities for musicians.

Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models

Training-free guidance enables controlled generation in diffusion and flow models, but most existing methods assume differentiable objectives and rely on gradients. This work focuses on training-free guidance addressing challenges from non-differentiable objectives and discrete data distributions. We propose an algorithmic framework TreeG: Tree Search-Based Path Steering Guidance, applicable to both continuous and discrete settings in diffusion and flow models. TreeG offers a unified perspective on training-free guidance: proposing candidates for the next step, evaluating candidates, and selecting the best to move forward, enhanced by a tree search mechanism over active paths or parallelizing exploration. We comprehensively investigate the design space of TreeG over the candidate proposal module and the evaluation function, instantiating TreeG into three novel algorithms. Our experiments show that TreeG consistently outperforms the top guidance baselines in symbolic music generation, small molecule generation, and enhancer DNA design, all of which involve non-differentiable challenges. Additionally, we identify an inference-time scaling law showing TreeG's scalability in inference-time computation.

MusicEval: A Generative Music Corpus with Expert Ratings for Automatic Text-to-Music Evaluation

The technology for generating music from textual descriptions has seen rapid advancements. However, evaluating text-to-music (TTM) systems remains a significant challenge, primarily due to the difficulty of balancing performance and cost with existing objective and subjective evaluation methods. In this paper, we propose an automatic assessment task for TTM models to align with human perception. To address the TTM evaluation challenges posed by the professional requirements of music evaluation and the complexity of the relationship between text and music, we collect MusicEval, the first generative music assessment dataset. This dataset contains 2,748 music clips generated by 31 advanced and widely used models in response to 384 text prompts, along with 13,740 ratings from 14 music experts. Furthermore, we design a CLAP-based assessment model built on this dataset, and our experimental results validate the feasibility of the proposed task, providing a valuable reference for future development in TTM evaluation. The dataset is available at https://www.aishelltech.com/AISHELL_7A.

TokenSynth: A Token-based Neural Synthesizer for Instrument Cloning and Text-to-Instrument

Recent advancements in neural audio codecs have enabled the use of tokenized audio representations in various audio generation tasks, such as text-to-speech, text-to-audio, and text-to-music generation. Leveraging this approach, we propose TokenSynth, a novel neural synthesizer that utilizes a decoder-only transformer to generate desired audio tokens from MIDI tokens and CLAP (Contrastive Language-Audio Pretraining) embedding, which has timbre-related information. Our model is capable of performing instrument cloning, text-to-instrument synthesis, and text-guided timbre manipulation without any fine-tuning. This flexibility enables diverse sound design and intuitive timbre control. We evaluated the quality of the synthesized audio, the timbral similarity between synthesized and target audio/text, and synthesis accuracy (i.e., how accurately it follows the input MIDI) using objective measures. TokenSynth demonstrates the potential of leveraging advanced neural audio codecs and transformers to create powerful and versatile neural synthesizers. The source code, model weights, and audio demos are available at: https://github.com/KyungsuKim42/tokensynth

TALKPLAY: Multimodal Music Recommendation with Large Language Models

We present TalkPlay, a multimodal music recommendation system that reformulates the recommendation task as large language model token generation. TalkPlay represents music through an expanded token vocabulary that encodes multiple modalities - audio, lyrics, metadata, semantic tags, and playlist co-occurrence. Using these rich representations, the model learns to generate recommendations through next-token prediction on music recommendation conversations, that requires learning the associations natural language query and response, as well as music items. In other words, the formulation transforms music recommendation into a natural language understanding task, where the model's ability to predict conversation tokens directly optimizes query-item relevance. Our approach eliminates traditional recommendation-dialogue pipeline complexity, enabling end-to-end learning of query-aware music recommendations. In the experiment, TalkPlay is successfully trained and outperforms baseline methods in various aspects, demonstrating strong context understanding as a conversational music recommender.

A Reservoir-based Model for Human-like Perception of Complex Rhythm Pattern

Rhythm is a fundamental aspect of human behaviour, present from infancy and deeply embedded in cultural practices. Rhythm anticipation is a spontaneous cognitive process that typically occurs before the onset of actual beats. While most research in both neuroscience and artificial intelligence has focused on metronome-based rhythm tasks, studies investigating the perception of complex musical rhythm patterns remain limited. To address this gap, we propose a hierarchical oscillator-based model to better understand the perception of complex musical rhythms in biological systems. The model consists of two types of coupled neurons that generate oscillations, with different layers tuned to respond to distinct perception levels. We evaluate the model using several representative rhythm patterns spanning the upper, middle, and lower bounds of human musical perception. Our findings demonstrate that, while maintaining a high degree of synchronization accuracy, the model exhibits human-like rhythmic behaviours. Additionally, the beta band neuronal activity in the model mirrors patterns observed in the human brain, further validating the biological plausibility of the approach.

GVMGen: A General Video-to-Music Generation Model with Hierarchical Attentions

Composing music for video is essential yet challenging, leading to a growing interest in automating music generation for video applications. Existing approaches often struggle to achieve robust music-video correspondence and generative diversity, primarily due to inadequate feature alignment methods and insufficient datasets. In this study, we present General Video-to-Music Generation model (GVMGen), designed for generating high-related music to the video input. Our model employs hierarchical attentions to extract and align video features with music in both spatial and temporal dimensions, ensuring the preservation of pertinent features while minimizing redundancy. Remarkably, our method is versatile, capable of generating multi-style music from different video inputs, even in zero-shot scenarios. We also propose an evaluation model along with two novel objective metrics for assessing video-music alignment. Additionally, we have compiled a large-scale dataset comprising diverse types of video-music pairs. Experimental results demonstrate that GVMGen surpasses previous models in terms of music-video correspondence, generative diversity, and application universality.

Diffusion based Text-to-Music Generation with Global and Local Text based Conditioning

Diffusion based Text-To-Music (TTM) models generate music corresponding to text descriptions. Typically UNet based diffusion models condition on text embeddings generated from a pre-trained large language model or from a cross-modality audio-language representation model. This work proposes a diffusion based TTM, in which the UNet is conditioned on both (i) a uni-modal language model (e.g., T5) via cross-attention and (ii) a cross-modal audio-language representation model (e.g., CLAP) via Feature-wise Linear Modulation (FiLM). The diffusion model is trained to exploit both a local text representation from the T5 and a global representation from the CLAP. Furthermore, we propose modifications that extract both global and local representations from the T5 through pooling mechanisms that we call mean pooling and self-attention pooling. This approach mitigates the need for an additional encoder (e.g., CLAP) to extract a global representation, thereby reducing the number of model parameters. Our results show that incorporating the CLAP global embeddings to the T5 local embeddings enhances text adherence (KL=1.47) compared to a baseline model solely relying on the T5 local embeddings (KL=1.54). Alternatively, extracting global text embeddings directly from the T5 local embeddings through the proposed mean pooling approach yields superior generation quality (FAD=1.89) while exhibiting marginally inferior text adherence (KL=1.51) against the model conditioned on both CLAP and T5 text embeddings (FAD=1.94 and KL=1.47). Our proposed solution is not only efficient but also compact in terms of the number of parameters required.